Ignition systems, and particularly transistor-controlled ignition systems are known. It has been previously proposed to vary the closed, or ON-TIME of the transistor switches independence on speed of the engine, so that the controlled transistor which controls opening and closing of the circuit through the primary of the ignition coil operates at its most efficient levels, just in saturation, but not remaining in saturated state for any appreciable period of time. If the supply voltage to the system varies widely-as is frequently the case in on-board automotive vehicle networks--it is difficult to maintain the operating efficiency of the final control transistor at optimum level, that is, to operate the control transistor so that it has just reached saturation, but does not remain in saturation, when it is controlled to be conductive. If the supply voltage is a battery then the ON-TIME, or closed time of the transistor switch must be so adjusted that the transistor will operate in saturation when the battery voltage is at its normal or design level, yet, that saturation conditions also pertain when the battery is close to being discharged. It is difficult to satisfy both requirements so that usually if the circuit is so arranged that, with fully charged battery the saturated condition of the transistor is properly obtained, the energy stored in the ignition coil when the battery is substantially discharged is decreased considerably.